DB2 disaster recovery high-availability improving method based on Itanium platform

The invention discloses a DB2 disaster recovery high-availability improving method based on an Itanium platform and belongs to a disaster recovery scheme. The technical scheme is as follows: databases are established by using shared storage by the aid of Inspur cluster management software K-HA, a K-HA server cluster is established, and main servers in the K-HA server cluster correspond to standby servers; when one main server system in the server cluster has a failure, server cluster resources are switched to another main server automatically; change data of the main servers are synchronously copied to the standby servers by the aid of the HADR characteristic of DB2; when the databases of the main servers have failures, the standby servers quickly take over loads; workloads of the databases are copied to the independent standby servers, so that when the main servers have local hardware failures or disastrous site failures, the workloads cannot be interrupted. According to the method, a high-reliability and high-availability database environment is established, and dual protection is performed on hardware layers and software layers of production databases respectively.